Customer Segmentation Using RFM Analysis and the K-Means Clustering Algorithm to Support Data-Driven Marketing Strategies

Rifat Naufal (Department of Computer Science, Universitas Negeri Semarang, Indonesia)
Alamsyah (Department of Computer Science, Universitas Negeri Semarang, Indonesia)

Information Technology Education Journal
Vol. 5, No. 3, August (2026)

Purpose – Customer segmentation plays an important role in supporting data-driven marketing decision-making. This study aimed to analyze and classify customers according to their buying pattern characteristics by implementing the RFM (Recency, Frequency, Monetary) technique in the K-Means Clustering algorithm.  Design/methods/approach– This study used the public Marketing Campaign dataset from Kaggle through several phases, covering data preprocessing, RFM model construction, logarithmic conversion, information standardization, identification of the most suitable number of clusters through the elbow technique and assessment through the Silhouette Score. Findings – The results showed that the most suitable number of clusters was k=3 with a Silhouette Score measurement of 0.503, indicating moderate clustering quality with acceptable cluster cohesion and separation. The resulting segmentation consisted of three main clusters, namely Loyal Customers, Need Attention Customers and At Risk Customers, where each cluster had different contribution characteristics and potential for the company. Research implications – The segmentation results provide practical recommendations that may help companies maintain Loyal Customers, improve engagement among Need Attention customers, and reduce customer churn risk through more targeted marketing strategies. This study used a single public dataset; therefore, the segmentation results may differ when applied to other datasets or industrial sectors. Originality/value – The originality of this study lies in the application of RFM-based customer value transformation prior to K-Means clustering on the Marketing Campaign dataset. This approach provides interpretable customer segmentation and behavioral insights that support data-driven marketing decision-making. The findings suggest that integrating RFM analysis with K-Means clustering can generate meaningful customer segments and provide actionable insights for customer retention, re-engagement, and loyalty management.